Isolation and Characterization of a Single-Stranded DNA Virus Infecting the Marine Planktonic Diatom Chaetoceros Sp

Plankton Benthos Res 7(1): 20–28, 2012 Plankton & Benthos Research © The Plankton Society of Japan

Isolation and characterization of a single-stranded DNA virus infecting the marine planktonic diatom Chaetoceros sp. (strain TG07-C28)

1,2 1 3 1 KENSUKE TOYODA , KEI KIMURA , NAOTSUGU HATA , NATSUKO NAKAYAMA , 1 1, KEIZO NAGASAKI & YUJI TOMARU *

1 National Research Institute of Fisheries and Environment of Inland Sea, Fisheries Research Agency, 2–17–5 Maruishi, Hatsukaichi, Hiroshima 739–0452, Japan 2 Present address: Research and Education Center for Natural Sciences, Keio University, 4–1–1 Hiyoshi, Kohoku-ku, Yokohama, Kanagawa 223–8521, Japan 3 Mie Prefecture Fisheries Research Institute, 3564–3 Hamajima, Hamajima, Shima, Mie 517–0404, Japan Received 19 August 2011; Accepted 1 December 2011

Abstract: Diatom dynamics in the ocean represents one of the most interesting research themes for marine ecolo- gists. Recent studies have shown the significance of viruses as potential sources of mortality for diatom populations, as well as the more traditional environmental factors. Here, we report the isolation and characterization of a new single-stranded DNA (ssDNA) virus (Csp05DNAV) that causes lysis in Chaetoceros sp. TG07-C28, isolated from the surface water of Ago Bay, Japan. The virus was isolated from the sediments in Ago Bay, and its infection was both lytic and species-specific. The latent period was estimated to be ＜24 h. The virion is 32–34 nm in diameter, and ac- cumulates in the nucleus of host cells. Csp05DNAV has a closed circular ssDNA genome, which includes a partially double-stranded region. Partial sequence analysis revealed that the open reading frame of this virus genome was similar to the putative replicase-related proteins of previously reported diatom viruses that harbor a ssDNA genome; specifically, CsalDNAV, CdebDNAV, ClorDNAV, and CtenDNAV. Based on the genome structure, Csp05DNAV is considered to belong to the genus Bacilladnavirus. As both the host and virus were isolated from Ago Bay, the host-virus system probably is ecologically important in the bay. This finding provides basic information towards elucidating diatom dynamics in coastal waters.

Key words: Bacilladnavirus, Chaetoceros, diatom, rod-shaped particle, ssDNA

Introduction While various environmental factors that are considered to affect diatom dynamics have been reported (Geider et al. Diatoms (Bacillariophyceae) are one of the most abun- 1988, Sarthou et al. 2005), recent studies have shown the dant groups of photosynthetic microorganisms in the potential significance of viruses in controlling the popula- ocean, and account for a large part of marine primary pro- tion dynamics of diatoms in natural environments (Tomaru duction (Nelson et al. 1995, Kooistra et al. 2007). Among et al. 2011a). this group, the genus Chaetoceros is highly diverse, with Diatom viruses are now represented by two recently ap- more than 400 species functioning as key primary produc- proved virus genera, Bacillarnavirus and Bacilladnavirus ers that sustain higher forms of aquatic life. In fact, Chaet- (International Committee on Taxonomy of Viruses, http:// oceros has been described as the most dominant phyto- talk.ictvonline.org/), and several uncategorized members. plankton group of the ocean (Rines & Hargraves 1988). The genera Bacillarnavirus and Bacilladnavirus comprise Therefore, within the framework of marine ecological diatom viruses with a linear single-stranded RNA (ssRNA) studies, it is important to understand the dynamics of dia- genome and a circular single-stranded DNA (ssDNA) ge- toms, including those belonging to the genus Chaetoceros. nome, respectively. Some diatom viruses have a genome type or structure that has not been sufficiently elucidated, * Corresponding author: Yuji Tomaru; E-mail, [email protected] and await categorization. All diatom viruses specifically ssDNA virus infecting Chaetoceros sp. 21

Ago Bay, Japan, on October 26, 2008, using an Ekman- Birge bottom sampler equipped with a 36 mm ϕ-corer tube (Yokoyama & Ueda 1997). Collected samples were deliv- ered to the laboratory without ﬁxation within 24 h of sam- pling, and stored at 4°C. Three grams of the sediment sam-

ple were shaken with 3 mL of N2SeO3-enriched SWM3 (400 rpm, 23°C, for 30 min), and centrifuged at 860×g at 4°C for 10 min. The supernatants were then passed through 0.2 μm Dismic-25cs filters (Advantec MFS Inc., Tokyo, Japan). Aliquots (0.2 mL) of the filtrates obtained from sediment samples were inoculated into exponentially-growing Chaetoceros sp. TG07-C28 cultures (0.8 mL), followed by incubation at 15°C using the light/dark cycle conditions de- Fig. 1. Chaetoceros sp. strain TG07-C28. (A) Optical micro- scribed above. Algal cultures inoculated with SWM3 graph of intact cells. (B) Optical micrograph of Csp05DNAV-in- served as controls. fected cells at 48 h post-inoculation. Bars indicate 100 mm. (C) From the cultures that showed an apparent crash in host Chaetoceros sp. cultures of control (left) and added Csp05DNAV cells after inoculation of the filtrates (e.g., Fig. 1), the re- (right), 4 days post-inoculation. sponsible pathogens were cloned through 2 extinction dilution cycles (Suttle 1993, Tomaru et al. 2004). Briefly, the algal lysate was diluted in modified SWM3 medium in a lyse their respective host diatom species; therefore, they series of 10-fold dilution steps. Aliquots (100 μL) of each are supposed to represent an important controlling agent dilution step were added to 8 wells in cell culture plates for the dynamics of their respective host populations. with BD Falcon 96 flat-bottom wells (Becton, Dickinson Although knowledge about the species and ecological and Company, Tokyo, Japan) containing 150 μL of an ex- characteristics of diatom viruses is gradually accumulat- ponentially growing host culture. Then, the algal lysate in ing, it is insufficient to understand the population dynam- the most diluted well of the first assay was carried over to ics of diatoms in natural waters. In the present report, we the second extinction dilution procedure. Finally, the resul- introduce a new circular ssDNA diatom virus that infects tant lysate in the final end-point dilution was used as a Chaetoceros sp. strain TG07-C28, where both the host and clonal lysate, in which the probability of two or more vi- the virus were isolated from Ago Bay, Japan. ruses occurring (i.e., failure in cloning) was estimated at ＜0.0106. Bacterial contamination was removed from each lysate in the highest dilution well of the second assay by Materials and Methods filtration through a 0.1 μm polycarbonate membrane filter (Whatman Ltd., Kent, UK), after which the lysate was Algal cultures and growth conditions transferred to another exponentially growing host culture. The axenic clonal algal strain used for virus isolation in To confirm or refute bacterial contamination, each lysate this study was Chaetoceros sp. strain TG07-C28 (Fig. 1). was observed using epifluorescence microscopy, after The strain was isolated from the surface water of Ago Bay staining with SYBR-Gold. Briefly, the lysate was fixed (34°17.643′N, 136°49.899′E), Japan, on July 4, 2007. Ago with glutaraldehyde at a final concentration of 1%, and Bay is located in a semi-enclosed area in a water depth of SYBR-Gold (Molecular Probes Inc., Eugene, OR, USA) ca. 10 m. Algal cultures were grown in modified SWM3 was added to each fixed sample at a final concentration of －4 medium that was enriched with 2 nM Na2SeO3 (Chen et al. 1.0×10 dilution of the commercial stock. The stained 1969, Itoh & Imai 1987) under a 12/12 h light-dark cycle of samples were filtered onto a 0.2 μm pore size polycarbon- ca. 110 to 150 μM of photons m－2 sec－1, using cool white ate membrane filter (Whatman Ltd., Kent, UK). Then, the fluorescent illumination at 15°C. The species of Chaetoc- filters were mounted on a glass slide with a drop of low- eros sp. strain TG07-C28 is considered to be different fluorescence immersion oil, and covered with another drop from the diatom host species of the viruses reported to of immersion oil and a cover slip. The slides were viewed date, according to analyses of PCR-restriction fragment at a magnification of ×1000 with an epifluorescence mi- length polymorphisms targeting the ribulose-1,5-bisphos- croscope (BX50, Olympus, Tokyo, Japan). The resultant phate carboxylase/oxygenase large subunit gene of chloro- axenic lysate was treated as a clonal virus suspension and plast DNA (Toyoda et al. 2011). The species to which this used for further analyis. strain belongs has not yet been determined based on its Host range morphological features. The inter-species host specificity of the isolated virus Virus isolation clone was tested by adding 5% (v/v) aliquots of fresh lysate Sediment samples (0–1 cm depth) were collected from that had been passed through a 0.2 μm pore size polycar- 22 K. Toyoda et al.

Table 1. Infection speciﬁcities of Csp05DNAV against 28 strains of marine phytoplankton.

Strains lysed by Famiy Species Strain code Temperature (°C) Csp05DNAVa

Bacillariophyceae Chaetoceros debilis Ch48 15 － Chaetoceros tenuissimus 2-10 15 － Chaetoceros salsugineum Ch42 15 － Chaetoceros socialis f. radians L-4 15 － Chaetoceros cf. affinis Ch5 15 － Chaetoceros lorenzianus ItDia-51 15 － Chaetoceros sp. TG07-C28 15 + Chaetoceros cf. pseudocurvisetus IT07-C37 15 － Detonula pumila IT09-K05 15 － Ditylum brightwellii IT09-K19 15 － Eucampia zodiacus EzB 15 － Rhizosolenia setigera S2 15 － Skeletonema sp. IT09-K17 15 － Stephanopyxis sp. IT09-K16 15 － Eustigmatophyceae Nannochloropsis sp. SFBB 20 － Cryptophyceae Teleaulax amphioxeia Tel5W 20 － Dinophyceae Alexandrium catenella ACNG 20 － Gymnodinium catenatum GC27-1 20 － Heterocapsa circularisquama HU9433-P 20 － Heterocapsa triquetra Ht 20 － Karenia mikimotoi GmH6 20 － Prorocentrum triestinum Pt-1 20 － Scrippsiella sp. SCKR 20 － Raphidophyceae Chattonella antiqua CaAR 20 － Chattonella marina CMKG-1 20 － Chattonella ovata CoV 20 － Fibrocapsa japonica F96 20 － Heterosigma akashiwo H93616 20 － a －, not lysed; +, lysed. bonate membrane filter (Whatman Ltd., Kent, UK) to du- concentration of 10% (w/v), and the suspension was stored plicate cultures of the following 28 exponentially growing at 4°C in the dark, overnight. After centrifugation at clonal algal strains that belong to the families of Bacillari- 57,000×g at 4°C for 1.5 h, the pellet was washed with ophyceae, Cryptophyceae, Dinophyceae, Eustigmatophy- 10 mM phosphate buffer (pH 7.2), and added to an equal ceae, and Raphidophyceae (shown in Table 1). The algal volume of chloroform. After vigorous vortexing, the sus- cultures were cultured under the conditions given above at pension was centrifuged at 2,200×g for 20 min at room either 15°C or 20°C. Growth, cell condition, and evidence temperature to remove the chloroform. The water phase of lysis in each algal culture were monitored by optical mi- was pipetted off and centrifuged at 217,000×g for 4 h at croscopy, and compared against control cultures that were 4°C to collect the virus particles. The resultant viral pellets inoculated with SWM3. Algal lysis was scored when an were used for genome analysis. The virus particles were aggregation of lysed cells was observed on the bottom of resuspended in 600 μL of ultrapure water, i.e., virus sus- the culture vessels. Cultures that did not appear to be lysed pension, for use in viral protein analysis and negative at 14 days post-inoculation (dpi) were scored as unsuitable staining observations under transmission electron micros- hosts for the viral pathogen. copy (TEM). Virus purification Viral proteins A 450 mL exponentially growing Chaetoceros sp. Aliquot (5 μL) of the virus suspension was mixed with 4 TG07-C28 culture was inoculated with 5 mL of the virus volumes of denaturing sample buffer (62.5 mM Tris-HCl suspension, and lysed. The lysate was passed through [pH 6.8], 5% 2-mercaptoethanol, 2% sodium dodecyl sul- 0.4 μm pore size polycarbonate membrane filters (What- fate [SDS], 20% glycerol, and 0.005% bromophenol blue), man Ltd., Kent, UK) to remove cellular debris. Polyethyl- and boiled for 5 min. The proteins were then separated by ene glycol 6000 (Wako Pure Chemical Industries Ltd., SDS-polyacrylamide gel electrophoresis (80×40×1 mm, Osaka, Japan) was added to the filtrate to obtain a final 12.5% polyacrylamide, 150 V) using the XV Pantera Sys- ssDNA virus infecting Chaetoceros sp. 23 tem (DRC Co. Ltd., Tokyo, Japan). Proteins were visual- using SYBR-Gold staining (Molecular Probes Inc., Eu- ized using Coomassie brilliant blue stain. Protein molecu- gene, OR, USA). lar mass standards (Bio-Rad Laboratories Inc., Hercules, Genome sequencing CA, USA) ranging from 10 to 250 kDa were used for size calibration. The sequencing of a partial viral genome was performed using the random amplified polymorphic DNA (RAPD) TEM method. In brief, RAPD-PCR amplification was conducted An exponentially growing culture of Chaetoceros sp. using a GeneAmp PCR System 9700 (Life Technologies TG07-C28 was inoculated with the virus suspension (5% Inc., Tokyo, Japan) with 50 μL mixtures containing v/v; multiplicity of virus infection, 100). As the control, a ＜500 ng viral template DNA, 1×KOD FX buffer (Toyobo Chaetoceros sp. TG07-C28 culture was inoculated with Co. Ltd., Osaka, Japan), each deoxynucleoside triphos- autoclaved culture medium SWM3. An aliquot of the cell phate (dNTP) at a concentration of 200 nM, 10 μM random suspension was sampled at 48 h post-inoculation (hpi), and deca nucleotides (BT set, Takara Bio Inc., Otsu, Japan), fixed with 2% glutaraldehyde and 3% paraformaldehyde in and 1 U of KOD FX DNA polymerase. The following cycle 0.1 M cacodylate buffer (pH 7.2) containing 2% NaCl for parameters were used for PCR: first, 5 rounds [denatur- 2 h at 4°C. Then, cells were collected by centrifugation, ation at 94°C (30 sec), annealing at 30°C (30 sec), and ex- washed with 0.1 M cacodylate buffer (pH 7.2) containing tension at 68°C (5 min)]; second, 5 rounds [denaturation at 2% NaCl for 2 h at 4°C, and embedded in 1% agarose, Type 94°C (30 sec), annealing at 35°C (30 sec), and extension at IX (Sigma-Aldrich Inc., St Louis, MO, USA). Washed 68°C (5 min)]; third, 5 rounds [denaturation at 94°C samples were post-fixed with 2% OsO4 in 0.1 M cacodylate (30 sec), annealing at 40°C (30 sec), and extension at 68°C buffer (pH 7.2) containing 2% NaCl. After washing with (5 min)]; fourth, 5 rounds [denaturation at 94°C (30 sec), buffer, the samples were dehydrated in a graded acetone annealing at 45°C (30 sec), and extension at 68°C (5 min)]; series, and then they were embedded in Spurr’s epoxy resin and fifth, 15 rounds [denaturation at 94°C (30 sec), anneal- on aluminum foil dishes. Samples were polymerized for ing at 50°C (30 sec), and extension at 68°C (5 min)]. The 12 h at 70°C. Seventy nm thin sections were cut using a di- PCR products were electrophoresed in 1% (w/v) Agarose amond knife on a Reichert Ultracut R microtome (Leica, ME gels (Wako Pure Chemical Industries Ltd., Osaka, Wetzlar, Germany), and mounted on formvar-coated one- Japan), in which the nucleic acids were visualized by ethid- slot grids. Sections were stained with 4% uranyl acetate ium bromide staining. The fragments at 0.5–3.0 kbp were and 3% lead citrate, and observed with a JEM-1010 elec- excised from the gel using Quantum Prep Freeze ’N tron microscope (JEOL Ltd., Tokyo, Japan). SqueezeTM DNA Gel Extraction Spin Columns (Bio-Rad The virus particles that were negatively stained with Laboratories Inc., Hercules, CA, USA), then purified using uranyl acetate were also observed using transmission elec- phenol-chloroform extraction, and dissolved in ultrapure tron microscopy. Briefly, a drop of purified virus suspen- water. The selected RAPD fragments were treated with sion was mounted on a grid (no. 780111630; JEOL Ltd., r-Taq DNA polymerase (Takara Bio Inc., Otsu, Japan) in Tokyo, Japan) for 30 sec, and excess water was removed the presence of dATP according to the manufacturer’s rec- using filter paper (no. 1; Advantec MFS Inc., Tokyo, ommendation, then incubated at 70°C for 1 h, and cloned Japan). Then, 4% uranyl acetate was applied for 10 sec, and using a TA cloning system (Life Technologies Inc., Tokyo, excess dye was removed using filter paper. After the grid Japan). PCR products were ligated into a TOPO pCRII was dried in a desiccator for ＞2 h, the negatively stained vector and transformed in Escherichia coli DH5a-compe- virus particles were observed using TEM at 80 kV. Particle tent cells (Toyobo Co. Ltd., Osaka, Japan). Sequencing was diameters were estimated using the negatively stained im- conducted using the dideoxy method with an ABI PRISM ages. 3100 Genetic Analyzer (Life Technologies Inc., Tokyo, Japan). The resultant fragment sequences were reassem- Viral nucleic acids bled using DNASTAR (DNASTAR Inc., WI, USA). Nucleic acids were extracted from the viral pellet using Southern blot analysis was conducted to distinguish the the DNeasy Mini Kit (Qiagen K.K., Tokyo, Japan). Fur- viral (+) and complementary (－) strand of the single- ther, aliquots (7 μL) of the nucleic acid solution were di- stranded region of the viral genomic DNA. On the basis of gested with RNase A (Nippon Gene Co. Ltd., Tokyo, the predicted partial sequence, digoxigenin-labelled RNA Japan) at 0.025 μg μL－1 for 1 h at 37°C, or incubated with probes specific for either the viral or complementary DNase I (Takara Bio Inc., Otsu, Japan) at 0.5 U μL－1 for 1 h strand were transcribed from the constructed plasmid with at 37°C, or with S1 nuclease (Takara Bio Inc., Otsu, Japan) T7 RNA polymerase or T3 RNA polymerase, respectively, at 0.7 U μL－1 for 15 min at 23°C. Nucleic acid extracts kept according to the manufacturer’s protocols (Promega K.K., on ice without treatment served as controls. The prepared Tokyo, Japan). The nucleotide sequence of the viral ge- nucleic acid samples were electrophoresed in agarose gels nome was determined by Southern dot-blot analysis using (1.2%; SeaKem® GTG Agarose, Lonza Inc., Basel, Swit- the probes according to a previously reported method zerland) at 50 V for 2.5 h. Nucleic acids were visualized (Mizumoto et al. 2007). The signals were detected with a 24 K. Toyoda et al. luminescence image analyzer (LAS-3000 mini, Fuji Photo Film, Tokyo, Japan). Putative open reading frames were identified using the ORF Finder (http://www.ncbi.nlm.nih.gov/gorf/gorf.html). Automated comparisons were conducted comparing the viral genome sequence using the BLAST program (Basic Local Alignment Research Tool, http://blast.ncbi.nlm.nih. gov/Blast.cgi) to genetic databases. Growth experiment An exponentially growing culture of Chaetoceros sp. strain TG07-C28 (50 mL) was inoculated at 15°C with the virus at a multiplicity of infection of 0.3. A Chaetoceros sp. strain TG07-C28 culture inoculated with an autoclaved culture medium served as the control. An aliquot of the cell suspension was sampled from each culture at 0, 12, 24, 30, 36, 48, 60, 72, 84, 96, and 108 h post-inoculation (hpi), and the number of host cells and viral infectious units was estimated. This experiment was a single trial. Cell counts were carried out with Fuchs-Rosenthal hemocytometer using optical microscopy (TE-300, Nikon, Tokyo, Japan), without fixation of the samples. The number of viral infectious units was determined using the extinction dilution Fig. 2. Major structural proteins of Csp05DNAV visualized by method (Suttle 1993). Briefly, the samples used to estimate sodium dodecyl sulfate-polyacrylamide gel electrophoresis. the viral infectious units were passed through 0.8 μm pore size polycarbonate membrane filters (Whatman Ltd., Kent, Host range UK) to remove cellular debris. These filtrates were diluted with modified SWM3 medium in a series of 10-fold dilu- The host range of the virus was tested using 28 phyto- tion steps. Aliquots (100 μL) of each dilution were added to plankton strains, including 14 diatom strains. The virus 8 wells in cell-culture plates with BD falcon 96 flat-bottom was lytic to its original host Chaetoceros sp. strain TG07- wells, and mixed with 150 μL of exponentially growing C28, but not to any other microalgal species that were culture of host algae. The cell culture plates were incu- tested (Table 1). These results show the high species-spe- bated at 15°C under a 12-h L : 12-h D cycle of 130 to cific infection of this virus and this is a general feature of 150 μM photons m－2 sec－1 with cool white fluorescent illu- microalgal viruses. mination, and were monitored over 14 days using optical Proteins microscopy (TE-300, Nikon, Tokyo, Japan) for the occur- rence of culture lysis. Due to the virus infection, the cul- The results of sodium dodecyl sulfate-polyacrylamide ture lysis was usually observed as almost complete crashes gel electrophoresis revealed two major protein bands that of the host cell population in a well. We calculated virus had a molecular weight of 75 and 40 kDa, respectively, and abundance from the number of wells in which algal lysis one faint band with a molecular weight of 86 kDa (Fig. 2). occurred using BASIC program (Nishihara et al. 1986). The gel electrophoresis pattern appeared to be different to The burst size was calculated by comparing the increase in that of other previously reported diatom DNA viruses, infectious titer and decrease in host cell number. such as CsalDNAV, CdebDNAV, and CtenDNAV, infecting Chaetoceros salsugineum Takano (Nagasaki et al. 2005), Chaetoceros debilis Cleve (Tomaru et al. 2008) and Chaet- Results and Discussion oceros tenuissimus Meunier (Tomaru et al. 2011b), respectively. The major protein bands of these three viruses are Isolation of the viral pathogen located between 22 and 46 kDa (Nagasaki et al. 2005, To- The viral isolate retained its lytic activity after filtration maru et al. 2008, Tomaru et al. 2011b). Furthermore, Clo- through a 0.2 μm filter. The lytic activities were serially rDNAV infecting Chaetoceros lorenzianus Grunow con- transferable to Chaetoceros sp. strain TG07-C28 cultures. tains just a single major polypeptide of 225 kDa (Tomaru et The cytoplasm and photosynthetic pigments of virus-in- al. 2011c), which differed to that of the virus isolated in the fected Chaetoceros sp. cells were degraded compared to present study. healthy cells (Fig. 1). Morphological features Thin sections of healthy Chaetoceros sp. strain TG07- ssDNA virus infecting Chaetoceros sp. 25

Fig. 3. Transmission electron micrographs of ultrathin sections of Chaetoceros sp. strain TG07-C28 and negatively stained Csp05DNAV particles. (A) Healthy cell. (B, C, D) Cells infected with Csp05DNAV at 48 h post-inoculation. (B) Virus-like particles (VLPs) are accumulated in the host nucleus. (C) Higher magniﬁcation of the VLPs in the host nucleus of panel B. (D) Higher magniﬁcation of the VLPs in the host nucleus. Arrows indicate rod-shaped particles. (E) Negatively stained Csp05D- NAV particles. Ch: chloroplast; M: mitochondrion; N: nucleus; V: virus-like particles.

C28 cells showed that the cytoplasmic organization and randomly assembled in the host nucleus (Fig. 3B, C and D). frustules are typical of these diatoms (Fig. 3A). In contrast, VLPs were not found in healthy control cells (Fig. 3A). electron micrographs of thin-sectioned cells at 48 h post-in- Furthermore, VLPs were also observed in culture lysates oculation (hpi) showed the presence of virus-like particles by negative staining electron microscopy (Fig. 3E). VLPs (VLPs) with a diameter of 32±2 nm (n=30), which were were hexagonal in outline, suggesting icosahedral symme- 26 K. Toyoda et al. try. They were 34±1 nm (n=30) in diameter, lacked a tail and outer membrane, and appeared similar to the VLPs observed in the host nucleus (Fig. 3C). Both the virion assem- blage site and particle diameter were similar to that of ssDNA diatom viruses, CtenDNAV (Tomaru et al. 2011b) and ClorDNAV (Tomaru et al. 2011c). Since (i) the algi- cidal pathogen was transferable to a fresh algal culture, (ii) VLPs were observed in the lysed culture, and (iii) VLPs were not found in healthy cultures, we concluded that the 32–34 nm particles observed within the infected cells and in the algal lysates comprised a highly plausible virus that is pathogenic to Chaetoceros sp. strain TG07-C28. To date, four different ssDNA viruses that infect Chaetoceros species have been reported; specifically, CsalDNAV, CdebDNAV, CtenDNAV, and ClorDNAV (Nagasaki et al. 2005, Tomaru et al. 2008, Tomaru et al. 2011b, Tomaru et al. 2011c). The virus identified in this study is the fifth case of an ssDNA virus infecting the genus Chaetoceros. This Fig. 4. Intact nucleic acids of Csp05DNAV (A), treatment with new virus was tentatively designated as Chaetoceros sp. DNase I (B, lane 1), RNase A (lane 2), S1 nuclease (lane 3). Sam- number 05 DNA virus (Csp05DNAV). ples were electrophoresed in an agarose gel. In addition to the VLPs of diameter 32 nm, rod-shaped particles of 19±1 nm (n=15) width were also observed in the host nucleus (Fig. 3D). Similar rod-shaped particles in DNA diatom viruses (Tomaru et al. 2011c). Based on these the virus-infected host cells have been reported for other data, we concluded that the viral genome consists of a sin- Chaetoceros viruses; specifically, CwNIV, CtenDNAV, gle strand of covalently closed circular DNA that is partly and ClorDNAV (Eissler et al. 2009, Tomaru et al. 2011b, double-stranded. The two minor bands are considered to Tomaru et al. 2011c). Yet, rod shaped particles have not be components of the virus genome; however, their fea- been observed in the viral lysate. Therefore, all of these tures were not revealed in this study. studies indicate that these particles in the virus-infected Partial sequencing of the Csp05DNAV genome and the host cells may be the precursors of mature virions. The Southern blot analysis identified 1,602 bases (AB647334) rod-shaped particles in the infected host cells were not ob- that included one open reading frame (ORF). This ORF served together with the icosahedral VLPs in the nega- contained 534 amino acids, and showed high similarity to tively stained viral suspensions, suggesting that the con- the putative replication-associated protein of ClorDNAV centration of rod-shaped particles might be considerably (E-value 2e－153), CsalDNAV (E-value 3e－137), CtenDNAV lower or absent in the lysates. However, the alternative hy- (E-value 3e－115), and CdebDNAV (E-value 1e－59). Csp05D- pothesis that the rod-shaped particle is a co-infecting virus NAV is considered to be a member of the genus Bacillad- cannot be excluded. Further analyses are necessary to clar- navirus, recently accepted as a genus by the International ify the role of these particles, and their relationships to the Committee on Taxonomy of Viruses, based on the present icosahedral virions. results, including the genome type, structure, and sequences. Other than these diatom viruses, no viruses in- Genomic analysis fecting marine micro-organisms were similar to Csp05D- The intact Csp05DNAV genome exhibited two major NAV. The ORF of Csp05DNAV showed low similarity bands (ca. 4.5 kb and 5.5 kb) and two minor bands (ca. with the replication protein of goose circovirus (E-value 0.6 kb and 1.0 kb) of nucleic acids (Fig. 4A). All bands were 1e－4) and the replication-associated protein of beak and sensitive to DNase I, but not to RNase A (Fig. 4B, lanes 1 feather disease virus (E-value 1e－4), both of which belong and 2, respectively); therefore, the viral genome is consid- to ssDNA viruses of the family Circoviridae and genus ered to be DNA. In addition, the genome was digested with Circovirus (Todd et al. 2000). S1 nuclease; however, a double-stranded DNA (dsDNA) of Replication ~0.9 kbp remained undigested (Fig. 4B, lane 3). In a pre- liminary PCR experiment using inverse PCR primer pairs, Chaetoceros sp. strain TG07-C28 grew exponentially the size of the amplicon was similar to that of the Csp05D- for 36 h in both control and virus-added cultures; however, NAV genome (unpublished data), which indicates that the the cell number in the inoculated culture rapidly decreased genome is a closed circular form (Tomaru et al. 2011b). after 48 h post-inoculation (hpi) (Fig. 5). The first signifi- These results are typical of a covalently closed circular cant increase in virus abundance was also observed at 24 ssDNA genome containing a partially double-stranded hpi (Fig. 5); thus, the latent period of Csp05DNAV ap- DNA region, as observed for previously reported circular peared to be ＜24 h. This discrepancy, whereby there is a ssDNA virus infecting Chaetoceros sp. 27

ously reported for the relationship between Heterocapsa circularisquama Horiguchi (Dinophyceae) and its infectious virus HcRNAV in Ago Bay (Tomaru et al. 2007). HcRNAV specifically increases its abundance in sediments during the host blooming periods, and survives within the sediment layer until the next bloom the following year. The temperature stability of the diatom ssDNA virus group is high (Tomaru et al. 2008, Tomaru et al. 2011b). In addition, Csp05DNAV was stably preserved for long periods of at least one year at temperatures below 4°C (unpublished data). In Ago Bay, the host-virus relationship between Chaetoceros sp. strain TG07-C28 and Csp05DNAV is likely to have been established and maintained over a long period of time. For further understanding of their ecological relationships in nature, developments of host species- specific detection and quantification methods such as using real-time PCR (Toyoda et al. 2010) are necessary, because the microscopic detection and abundance determination of Chaetoceros sp. strain TG07-C28 in natural waters will be considerably difficult. The relationship between Chaetoceros dynamics and its Fig. 5. Changes in cell number of Chaetoceros sp. strain infectious viruses in natural environments is being gradu- TG07-C28 used for growth experiments with (■) or without ally revealed. The diatom viruses in the water column rap- (□) virus Csp05DNAV inoculation, and the virus titer (●). The idly increase during their host diatom blooms, C. tenuissi- number of host cells and viruses were estimated by direct count- mus and C. debilis, and are maintained at a high abun- ing using microscopy and the extinction dilution method, respec- dance throughout the blooming period (Tomaru et al. tively. Virus inoculation was performed at 0 days in an exponen- 2011a). Diatom populations appear to be affected by their tially growing host culture with a multiplicity of infection of 0.3. infectious viruses, even when both host and virus concen- trations are relatively low (Tomaru et al. 2011a). Recent studies have also shown that two distinctive viruses (i.e., parallel increase in host and virus numbers, is commonly CtenRNAV and CtenDNAV) can share the same diatom observed in diatom host-virus systems (Shirai et al. 2008, species C. tenuissimus and might simultaneously affect the Tomaru et al. 2011b). This is considered to occur due to a host population dynamics (Tomaru et al. 2011b). These ob- low percentage of virus sensitive cells being present during servations suggest that diatom viruses affect their respec- the logarithmic growth phase of the host populations, the tive host population dynamics in natural environments. numbers of which might increase during a stationary The genus Chaetoceros is one of the major phytoplank- phase. ters in the ocean and plays an important role as primary The average hosts/virus ratio at 36 to 60 hpi was used to producers (Rines & Hargraves 1988). This genus includes calculate the burst size, which was estimated to be 4.3× at least 400 species, and blooms of Chaetoceros are often 102 infectious units cell－1. The burst sizes of the previously composed of multiple species, with one example exceeding reported ssDNA diatom viruses range between ~101–102 15 species (Rines & Hargraves 1988). Many recent reports infectious units cell－1 except for that of ClorDNAV, at ~104 on the isolation of Chaetoceros viruses, e.g. reviewed in infectious units cell－1 (Tomaru & Nagasaki 2011, Tomaru Tomaru & Nagasaki (2011), suggest that the population dy- et al. 2011b, 2011c). The diversities of the burst sizes might namics of each species might be affected by its infectious reflect the differences of the ecological strategies for each viruses. To reveal further the dynamics of Chaetoceros ssDNA virus species. The burst size of the viruses should populations and associated viral effects, the isolation and be determined under various environmental conditions, characterization of new diatom viruses should be contin- e.g. nutrients, salinity and temperature, to reveal the ef- ued. fects of the diatom viruses on the host population dynamics in future studies. Acknowledgements Ecological implications This study was supported by Grants-in-Aid for Young Both the host species and the virus Csp05DNAV were Scientists (A) (22688016) from the Ministry of Education, isolated from Ago Bay, Japan. Because the virus was iso- Science, and Culture of Japan. lated from the sediments, the bottom sediments may serve as a reservoir for the virus throughout the year, as previ- 28 K. Toyoda et al.

book of Methods in Aquatic Microbial Ecology (eds Kemp PF, References Sherr E, Cole JJ). Lewis Publishers, Boca Raton, pp. 121–137. Chen LCM, Edelstein T, McLachlan J (1969) Bonnemaisonia Todd D, McNulty MS, Mankertz A, Lukert P, Randels JW, Dale hamifera Hariot in nature and in culture. J Phycol 5: 211–220. JL (2000) Family Circoviridae. In: Virus Taxonomy, Classifi- Eissler Y, Wang K, Chen F, Wommack E, Coats W (2009) Ultra- cation, and Nomenclature of Viruses, 7th report (eds Van Re- structural characterization of the lytic cycle of an intranuclear genmortel MHV, Fauquet CM, Bishop DHL, Carsten EB, virus infecting the diatom Chaetoceros cf. wighamii (bacillari- Estes MK, Lemon SM, Maniloff J, Mayo MA, McGeoch DJ, ophyceae) from Chesapeake Bay, USA. J Phycol 45: 787–797. Pringle CR, Wickner RB). Academic Press, San Diego, pp. Geider RJ, MacIntyre HL, Kana TM (1988) A dynamic regula- 299–303. tory model of phytoplanktonic acclimation to light, nutrients, Tomaru Y, Fujii N, Oda S, Toyoda K, Nagasaki K (2011a) Dy- and temperature. Limnol Oceanogr 43: 679–694. namics of diatom viruses on the western coast of Japan. Aquat Itoh K, Imai I (1987) Rafido so [Raphidophyceae]. In: A Guide Microb Ecol 63: 223–230. for Studies of Red Tide Organisms (ed Japan Fisheries Re- Tomaru Y, Hata N, Masuda T, Tsuji M, Igata K, Masuda Y, Yam- source Conservation Association). Shuwa, Tokyo, pp. 122– atogi T, Sakaguchi M, Nagasaki K (2007) Ecological dynam- 130. (in Japanese) ics of the bivalve-killing dinoflagellate Heterocapsa circular- Kooistra WHCF, Gersonde R, Medlin LK, Mann DG (2007) The isquama and its infectious viruses in different locations of origin and evolution of the diatoms: Their adaptation to a western Japan. Environ Microbiol 9: 1376–1383. planktonic existence. In: Evolution of Primary Producers in Tomaru Y, Nagasaki K (2011) Diatom viruses. In: The Diatom the Sea (eds Falkowski PG, Knoll A). Elsevier, San Diego, pp. World, Cellular Origin, Life in Extreme Habitats and Astrobi- 207–249. ology vol 19 (eds Seckbach J, Kociolek JP). Springer, London, Mizumoto H, Tomaru Y, Takao Y, Shirai Y, Nagasaki K (2007) pp. 211–225. Intraspecies host specificity of a single-stranded RNA virus Tomaru Y, Shirai Y, Suzuki H, Nagumo T, Nagasaki K (2008) infecting a marine photosynthetic protist is determined at the Isolation and characterization of a new single-stranded DNA early steps of infection. J Virol 81: 1372–1378. virus infecting the cosmopolitan marine diatom Chaetoceros Nagasaki K, Tomaru Y, Takao Y, Nishida K, Shirai Y, Suzuki H, debilis. Aquat Microb Ecol 50: 103–112. Nagumo T (2005) Previously unknown virus infects marine Tomaru Y, Shirai Y, Toyoda K, Nagasaki K (2011b) Isolation and diatom. Appl Environ Microbiol 71: 3528–3535. characterisation of a single-stranded DNA virus infecting the Nelson DM, Treguer P, Brzezinski MA, Leynaert A, Queguiner marine planktonic diatom Chaetoceros tenuissimus Meunier. B (1995) Production and dissolution of biogenic silica in the Aquat Microb Ecol 64: 175–184. ocean: Revised global estimates, comparison with regional Tomaru Y, Takao Y, Suzuki H, Nagumo T, Koike K, Nagasaki K data and relationship to biogenic sedimentation. Global Bio- (2011c) Isolation and characterization of a single-stranded geochem Cycles 9: 359–372. DNA virus infecting Chaetoceros lorenzianus Grunow. Appl Nishihara T, Kurano N, Shinoda S (1986) Calculation of most Environ Microbiol 77: 5285–5293. probable number for enumeration of bacteria on microcom- Tomaru Y, Tarutani K, Yamaguchi M, Nagasaki K (2004) Quan- puter. Eisei Kagaku 32: 226–228. (in Japanese with English titative and qualitative impacts of viral infection on Hetero- abstract) sigma akashiwo (Raphidophyceae) population during a bloom Rines JBE, Hargraves PE (1988) The Chaetoceros Ehrenberg in Hiroshima Bay, Japan. Aquat Microb Ecol 34: 227–238. (Bacillariophyceae) flora of Narragansett Bay, Rhode Island, Toyoda K, Nagasaki K, Tomaru Y (2010) Application of real- USA. Bibl Phycol 79: 1–196. time PCR assay for detection and quantification of bloom- Sarthou G, Timmermans KR, Blain S, Treguer P (2005) Growth forming diatom Chaetoceros tenuissimus Meunier. Plankton physiology and fate of diatoms in the ocean: a review. J Sea Benthos Res 5: 56–61. Res 53: 25–42. Toyoda K, Nagasaki K, Williams DM, Tomaru Y (2011) PCR- Shirai Y, Tomaru Y, Takao Y, Suzuki H, Nagumo T, Nagasaki K RFLP analysis for species-level distinction of the genus Chaet- (2008) Isolation and characterization of a single-stranded oceros Ehrenberg (Bacillariophyceae). Hiyoshi Rev Natur Sci RNA virus infecting the marine planktonic diatom Chaeto- 50: 21–30. ceros tenuissimus Meunier. Appl Environ Microbiol 74: 4022– Yokoyama K, Ueda H (1997) A simple corer set inside an Ekman 4027. grab to sample intact sediments with the overlying water. Ben- Suttle CA (1993) Enumeration and isolation of viruses. In: Hand- thos Res 52: 119–122.